Converting biennial data to yearly (panel)

[Ijaz AK]

What is the easiest way to convert a biennial data to yearly data? I have a biennial data for each of the 50 states from 1976-2022. But I want to make it yearly for each state.

* Example generated by -dataex-. For more info, type help dataex
clear
input int year str20 state float voted
1976 "ALABAMA"     984181
1978 "ALABAMA"     642279
1980 "ALABAMA"    1013626
1982 "ALABAMA"     961019
1984 "ALABAMA"    1148574
1986 "ALABAMA"    1115517
1988 "ALABAMA"    1178298
1990 "ALABAMA"    1015869
1992 "ALABAMA"    1602536
1994 "ALABAMA"    1115019
1996 "ALABAMA"    1468693
1998 "ALABAMA"    1215179
2000 "ALABAMA"    1438994
2002 "ALABAMA"    1268802
2004 "ALABAMA"    1792759
2006 "ALABAMA"    1140152
2008 "ALABAMA"    1855268
2010 "ALABAMA"    1367747
2012 "ALABAMA"    1933630
2014 "ALABAMA"    1080880
2016 "ALABAMA"    1889685
2018 "ALABAMA"    1659895
2020 "ALABAMA"    2051659
2022 "ALABAMA"    1343710
1976 "ALASKA"      118208
1978 "ALASKA"      124187
1980 "ALASKA"      154618
1982 "ALASKA"      181084
1984 "ALASKA"      206437
1986 "ALASKA"      180277
1988 "ALASKA"      192955
1990 "ALASKA"      191647
1992 "ALASKA"      239116
1994 "ALASKA"      208240
1996 "ALASKA"      233700
1998 "ALASKA"      223300
2000 "ALASKA"      274393
2002 "ALASKA"      227725
2004 "ALASKA"      299996
2006 "ALASKA"      234645
2008 "ALASKA"      316978
2010 "ALASKA"      254335
2012 "ALASKA"      289804
2014 "ALASKA"      279741
2016 "ALASKA"      308198
2018 "ALASKA"      282166
2020 "ALASKA"      353165
2022 "ALASKA"      263610
end

[Andrew Musau]

No need to. Just let Stata know that you have biennial data.

encode state, gen(State)
xtset State year, delta(2)

Res.:

. encode state, gen(State)

.
. xtset State year, delta(2)

Panel variable: State (strongly balanced)
 Time variable: year, 1976 to 2022
         Delta: 2 units

[Nick Cox]

Cross-posted at https://www.reddit.com/r/stata/comme...o_yearly_data/

Please note that it is a request here and a rule on that part of Reddit to tell people about cross-posting.

The replies on Reddit range from puzzled to bizarre. A suggestion there to reshape wide is hard to understand.

I support Andrew Musau's suggestion above. You can't get yearly data out of this dataset without interpolation, which in this context is probably a bad idea on several grounds. Were there votes in odd years and what sense does it make to impute or interpolate?

Perhaps you have other variables for every year, but if so we need to know more about your goals to add to Andrew's advice.